| Popis: |
The study presents the finding that the bonus program raised math achievement in schools with a small number of teachers with tested students, although these impacts are small and only marginally significant in the program’s second year (the effect in the first year is the finding chosen for the SCORE program). This reflects the following statement from the paper's abstract: "Although overall the program had little effect on student achievement, we show that in schools where incentives to free ride were weakest, the program led to small increases in math achievement." The claim is tested with an analysis of a bonus program in New York City schools. In 2007, close to 200 schools were randomly selected from a group of high-poverty schools. These schools could earn school-wide bonuses by surpassing goals primarily based on student achievement. The dependent variable--average math test scores--is regressed on an indicator for selection into the bonus program’s treatment group, the interaction of the treatment indicator with an indicator for schools in the bottom quartile of the number of teachers with tested students, and controls (average math score in 2007; school type indicators; percentage of students who are English Language Learners, in special education, Title 1 free lunch recipients, and minorities; and performance under the New York City accountability system). The model supplying evidence for the SCORE program is the Year 1 (2007-2008) effects in Model 6 of Table 1, and the effect of interest is the treatment effect for schools in the first quartile of number of teachers with tested students. For schools at the bottom of the distribution of the number of teachers with tested students, the authors estimate a positive effect of the bonus program on math achievement in the first year. In 2008, the bonus program resulted in a 3.2-point (0.08 student-level standard deviation) increase in math achievement (treatment effect estimate = 3.225, robust SE = 1.395, p < .05). The focal test in the original paper is: treatment effect estimate = 3.225, robust SE = 1.395, p < .05 |
